Collaborative Research: SaTC: CORE: Large: Privacy-Preserving Abuse Prevention for Encrypted Communications Platforms

协作研究：SaTC：核心：大型：加密通信平台的隐私保护滥用预防

• 批准号: 2120497
• 负责人: Amy Zhang
• 金额: $ 37.14万
• 依托单位: University of Washington
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-08-01 至 2026-07-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2120497&HistoricalAwards=false
• 关键词: Collaborative; Research; SaTC; CORE; Large

This project will address the immense challenge of mitigating abuse in communication services where interactions between the parties are private and fully encrypted. Such services have become popular for individual and group communications, and their security features help protect individual privacy and rights. But these platforms are also used for harmful and illegal purposes such as organizing violent activities or sharing child sexual abuse materials. In addition, users on these platforms are subject to abuse such as hate, harassment, and misinformation. The encryption used by such services makes detecting and blocking harmful content extremely difficult. This work will develop new trust and safety approaches to enable secure and trustworthy communications that preserve privacy while mitigating abuses. The project will aim to provide (1) technical advances in developing novel cryptographic tools and techniques to support mitigation of abuse; (2) human-centered advances in understanding perceptions and expectations of privacy and abuse mitigation, as well as creating novel designs for individual and community interactions; and (3) legal, policy, and regulatory advances to support and enable these abuse-mitigating features.The research effort is organized around two overlapping thrusts: algorithmic-driven approaches and community-driven approaches. The algorithmic approaches will focus on developing better cryptographic tools for privacy-aware abuse detection in encrypted settings, such as detection of viral, fast-spreading content. These designs will be informed by a human-centered approach to understanding people's privacy expectations, and supported by legal analyses that ensure tools are consistent with applicable privacy and content-moderation laws. In the second thrust, the community approaches will focus on providing communities with the tools they need to address abuse challenges in encrypted settings. Given the challenges and pitfalls of centralized approaches for abuse mitigation, the project will explore building distributed capabilities to support communities and groups on these platforms. An importantly ingredient is working with communities and community moderators to understand their needs, as well as guide design of legal and policy frameworks to support new approaches.Taken together, this project will address the challenge of abuse mitigation on encrypted platforms, while preserving privacy protections for individuals and communities. It will especially consider the perspectives of individuals and communities most in need of privacy and abuse protection. The work, if successful, should fuel new innovations in the design of encrypted messaging platforms, and in basic research in cryptography, human-centered design, and Internet law.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

该项目将解决减少通信服务滥用的巨大挑战，其中各方之间的交互是私密且完全加密的。此类服务已在个人和团体通信中变得流行，其安全功能有助于保护个人隐私和权利。但这些平台也被用于有害和非法目的，例如组织暴力活动或分享儿童性虐待材料。此外，这些平台上的用户还容易受到仇恨、骚扰和错误信息等虐待。此类服务使用的加密使得检测和阻止有害内容变得极其困难。这项工作将开发新的信任和安全方法，以实现安全且值得信赖的通信，在保护隐私的同时减少滥用行为。该项目旨在提供 (1) 开发新型加密工具和技术方面的技术进步，以支持减轻滥用行为； (2) 在理解隐私和滥用缓解的看法和期望方面取得以人为本的进步，并为个人和社区互动创造新颖的设计； (3) 法律、政策和监管方面的进步，以支持和实现这些减少滥用的功能。研究工作围绕两个重叠的主旨进行组织：算法驱动的方法和社区驱动的方法。 算法方法将侧重于开发更好的加密工具，用于加密环境中的隐私意识滥用检测，例如检测病毒式、快速传播的内容。这些设计将以以人为本的方法来了解人们的隐私期望，并得到法律分析的支持，以确保工具符合适用的隐私和内容审核法律。在第二个重点中，社区方法将侧重于为社区提供解决加密环境中滥用挑战所需的工具。考虑到集中式缓解滥用方法的挑战和陷阱，该项目将探索构建分布式功能来支持这些平台上的社区和团体。一个重要的组成部分是与社区和社区管理员合作，了解他们的需求，并指导法律和政策框架的设计以支持新方法。总之，该项目将解决加密平台上滥用缓解的挑战，同时保留隐私保护对于个人和社区。它将特别考虑最需要隐私和滥用保护的个人和社区的观点。这项工作如果成功，将推动加密消息平台的设计以及密码学、以人为本的设计和互联网法的基础研究方面的新创新。该奖项反映了 NSF 的法定使命，并通过使用评估结果被认为值得支持。基金会的智力价值和更广泛的影响审查标准。

项目成果

Skin Deep: Investigating Subjectivity in Skin Tone Annotations for Computer Vision Benchmark Datasets

肤浅：研究计算机视觉基准数据集肤色注释的主观性

• DOI: 10.1145/3593013.3594114
• 发表时间: 2023-05-15
• 期刊: Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency
• 作者: Teanna Barrett;Quan Ze Chen;Amy X. Zhang
• 通讯作者: Amy X. Zhang